English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

In the whole aircraft structural optimization loop, thermal analysis plays avery important role. But it faces a severe computational burden when directlyapplying traditional numerical analysis tools, especially when eachoptimization involves repetitive parameter modification and thermal analysisfollowed. Recently, with the fast development of deep learning, severalConvolutional Neural Network (CNN) surrogate models have been introduced toovercome this obstacle. However, for temperature field prediction on irregulargeometric domains (TFP-IGD), CNN can hardly be competent since most of themstem from processing for regular images. To alleviate this difficulty, wepropose a novel physics and data co-driven surrogate modeling method. First,after adapting the Bezier curve in geometric parameterization, a body-fittedcoordinate mapping is introduced to generate coordinate transforms between theirregular physical plane and regular computational plane. Second, aphysics-driven CNN surrogate with partial differential equation (PDE) residualsas a loss function is utilized for fast meshing (meshing surrogate); then, wepresent a data-driven surrogate model based on the multi-level reduced-ordermethod, aiming to learn solutions of temperature field in the above regularcomputational plane (thermal surrogate). Finally, combining the grid positioninformation provided by the meshing surrogate with the scalar temperature fieldinformation provided by the thermal surrogate (combined model), we reach anend-to-end surrogate model from geometric parameters to temperature fieldprediction on an irregular geometric domain. Numerical results demonstrate thatour method can significantly improve accuracy prediction on a smaller datasetwhile reducing the training time when compared with other CNN methods.

A physics and data co-driven surrogate modeling approach for temperature field prediction on irregular geometric domain